CN110751949A

CN110751949A - Voice recognition method and device and computer readable storage medium

Info

Publication number: CN110751949A
Application number: CN201910997985.2A
Authority: CN
Inventors: 高均波; 陈孝良; 常乐
Original assignee: Beijing Sound Intelligence Technology Co Ltd
Current assignee: Beijing Sound Intelligence Technology Co Ltd; Beijing SoundAI Technology Co Ltd
Priority date: 2019-10-18
Filing date: 2019-10-18
Publication date: 2020-02-04

Abstract

The present disclosure provides a speech recognition method, comprising: monitoring pickup sound intensity data, and judging a working scene according to the pickup sound intensity data to obtain a working scene judgment result; selecting audio data of at least one first microphone or at least one second microphone according to the working scene judgment result, wherein a first parameter of the first microphone is within a first preset range, and a second parameter of the second microphone is within a second preset range; and performing voice recognition by using the audio data of the selected microphone. And through the judgment of the working scene, the data sent by the devices with different parameter indexes are selected for voice recognition, so that the audio acquisition circuit is suitable for different working scenes.

Description

Voice recognition method and device and computer readable storage medium

Technical Field

The present disclosure relates to the field of audio circuits, and in particular, to a method and an apparatus for speech recognition and a computer-readable storage medium.

Background

The existing audio acquisition circuit is commonly used in different working scenes, and has different index requirements in different working scenes, but due to the limitation of device performance, the requirements of all working scenes are hardly considered. For example, in a conventional microphone array circuit, due to the influence of the performance of a microphone device, either the AOP (Acoustic Overload Point) index is generally not applicable to an environment with particularly severe Noise, or the AOP index is excellent, but the SNR (Signal to Noise Ratio) index is not good, and the sound pickup quality is poor. Therefore, the existing microphone array is either special for the common environment and has higher SNR performance; or the method is used in a particularly severe noise environment, has a high AOP index and cannot give consideration to both the noise and the AOP index.

Disclosure of Invention

Technical problem to be solved

The present disclosure provides a speech recognition method, apparatus and computer-readable storage medium to at least partially solve the above-mentioned technical problems.

(II) technical scheme

According to an aspect of the present disclosure, there is provided a speech recognition method including:

monitoring pickup sound intensity data, and judging a working scene according to the pickup sound intensity data to obtain a working scene judgment result;

selecting first audio data of at least one first microphone or second audio data of at least one second microphone according to the working scene judgment result, wherein a first parameter of the first microphone is within a first preset range, and a second parameter of the second microphone is within a second preset range;

and performing voice recognition by using the selected first audio data or the selected second audio data.

In some embodiments, the first parameter AOP of the first microphone is higher than a first preset value and the second parameter SNR of the second microphone is higher than a second preset value.

In some embodiments, the selecting the first audio data of the at least one first microphone or the second audio data of the at least one second microphone comprises:

transmitting a gating signal, gating a data line and/or a power supply of at least one first microphone and/or at least one second microphone, and receiving audio data of the gated microphone; or

The received first audio data of the at least one first microphone and the second audio data of the at least one second microphone are selected.

In some embodiments, the performing a work scene determination according to the pickup sound intensity data to obtain a work scene determination result includes:

comparing the monitored pickup sound intensity data with a preset switching threshold;

if the pickup sound intensity data is larger than the preset switching threshold value, judging that the current scene is a first working scene suitable for the first microphone;

and if the pickup sound intensity data is smaller than or equal to the preset switching threshold value, judging that the current scene is a second working scene suitable for the second microphone.

In some embodiments, the initial value of the working scenario determination result defaults to a second working scenario in which the second microphone is applied, or a first working scenario in which the first microphone is applied.

In some embodiments, the selecting, according to the work scenario determination result, first audio data of at least one first microphone or second audio data of at least one second microphone includes:

if the working scene judgment result is a first working scene, sending a gating control signal, gating at least one first microphone, and receiving first audio data of the at least one first microphone for voice recognition;

and if the working scene judgment result is a second working scene, sending a gating control signal, gating the at least one second microphone, and receiving second audio data of the at least one second microphone for voice recognition.

In some embodiments, the selecting, according to the work scenario determination result, the first audio data of the at least one first microphone or the second audio data of the at least one second microphone includes:

receiving at least one first microphone and at least one second microphone, and buffering and storing first audio data of the at least one first microphone and second audio data of the at least one second microphone;

and selecting the first audio data or the second audio data for voice recognition according to the working scene judgment result.

In some embodiments, the speech recognition method further comprises:

selecting the first audio data or the second audio data for voice recognition according to the working scene judgment result, judging whether the voice recognition result is correct or not after the voice recognition result is obtained, and taking the voice recognition result as a final voice recognition result if the voice recognition result is the correct result; otherwise, the other group of audio data is adopted for voice recognition, and the two voice recognition processing results are synthesized for judgment to obtain the final voice recognition result.

In some embodiments, the speech recognition method further comprises:

and when one group of the at least one first audio data or the at least one second audio data is selected to perform voice recognition according to the working scene judgment result, continuously monitoring pickup sound intensity data and performing working scene judgment, and if the working scene judgment result changes, performing voice recognition by adopting the other group of audio data.

According to another aspect of the present disclosure, there is provided a voice recognition apparatus including:

the mode selection preprocessing unit is used for monitoring pickup sound intensity data and judging a working scene according to the pickup sound intensity data to obtain a working scene judgment result;

the working mode selection unit is connected to the mode selection preprocessing unit and used for selecting first audio data of at least one first microphone or second audio data of at least one second microphone according to the working scene judgment result, wherein the AOP of the first microphone is higher than a first preset value, and the SNR of the second microphone is higher than a second preset value;

and the voice recognition processing unit is connected to the working mode selection unit and used for carrying out voice recognition on the data of the gated microphone.

In some embodiments, the mode selection pre-processing unit further comprises:

the comparison unit is used for comparing the monitored pickup sound intensity data with a preset switching threshold value, and if the pickup sound intensity data is larger than the preset switching threshold value, judging that the current scene is a first working scene suitable for the first microphone; and if the pickup sound intensity data is smaller than or equal to the preset switching threshold value, judging that the current scene is a second working scene suitable for the second microphone.

In some embodiments, the speech recognition processing unit further comprises:

and the voice awakening unit is used for voice awakening before voice recognition.

According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the speech recognition method as set forth above.

(III) advantageous effects

According to the technical scheme, the method has the following beneficial effects that:

through the judgment of the working scene, the data sent by the devices with different parameter indexes are selected for voice recognition, so that the audio acquisition circuit is suitable for different working scenes, for example, the switching between microphones with high SNR and high AOP is realized, and the audio acquisition circuit is suitable for the working environment of common working environment and particularly severe noise environment.

Drawings

Fig. 1A is a schematic structural diagram of a microphone array according to a first embodiment of the disclosure.

Fig. 1B is a diagram of a microphone data line strobe circuit according to a first embodiment of the present disclosure.

Fig. 2 is a circuit diagram of a data line gating circuit of a microphone according to a second embodiment of the present disclosure.

Fig. 3 is a circuit configuration diagram of a power gating circuit of a microphone according to a third embodiment of the present disclosure.

Fig. 4 is a circuit configuration diagram of a power gating circuit of a microphone according to a fourth embodiment of the present disclosure.

Fig. 5 is a flowchart of a speech recognition method according to a sixth embodiment of the present disclosure.

Fig. 6 is a flowchart of a method for not performing mode switching in a single voice interaction process according to a sixth embodiment of the present disclosure.

Fig. 7 is a flowchart of a method for mode switching during a single voice interaction process according to a sixth embodiment of the present disclosure.

Fig. 8 is a flowchart of a speech recognition method according to a sixth embodiment of the present disclosure, which integrates results of speech recognition processing.

Fig. 9 is a block diagram of a speech recognition apparatus according to a seventh embodiment of the present disclosure.

Detailed Description

The present disclosure provides a speech recognition method and apparatus, which can be applied to different working scenes by adaptively switching devices with different parameter indexes through a gating circuit. Particularly, the gating method of the microphone array is provided, so that the switching between the microphones with high SNR and high AOP is realized, and the method is suitable for the working environment of common working environment and particularly severe noise environment.

For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.

Certain embodiments of the present disclosure will now be described more fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the disclosure are shown. Indeed, various embodiments of the disclosure may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements.

In a first exemplary embodiment of the present disclosure, a gating circuit is provided. The gating circuit is connected with at least one first electric appliance and at least one second electric appliance, a first parameter of the first electric appliance is within a first preset range, and a second parameter of the second electric appliance is within a second preset range. In this embodiment, the first electrical appliance is a first microphone, and the second electrical appliance is a second microphone.

Fig. 1A is a schematic structural diagram of a microphone array according to a first embodiment of the disclosure. As shown in fig. 1A, at each microphone location in the microphone array, two microphones are used, one using a high AOP microphone and the other using a high SNR microphone.

Referring back to fig. 1, the first microphone MIC11 is a high AOP microphone, the second microphone MIC12 is a high SNR microphone, the MIC21 is a high AOP microphone, the MIC22 is a high SNR microphone, and so on.

In this embodiment, the first microphone MIC11 and the second microphone MIC12 are gated by a gating circuit. The gating circuit is a microphone data line gating circuit and is used for gating the data line of the at least one first microphone or the at least one second microphone. Specifically, the microphone data line gating circuit comprises a first switch circuit and a second switch circuit.

The first switch circuit is connected with a first microphone, the input end of the first switch circuit is connected with a first gating control signal, and the output end of the first switch circuit is a first microphone data output end; the first switching circuit comprises a first switchable device, and the first switchable device is connected to a first microphone and a first gating control signal;

the second switch circuit is connected with a second microphone, the input end of the second switch circuit is connected with a second gating control signal, and the output end of the second switch circuit is a second microphone data output end; the second switching circuit includes a second switchable device connected to a second microphone and a second gating control signal.

Fig. 1B is a circuit diagram of a microphone data line gating circuit according to a first embodiment of the present disclosure. As shown in fig. 1B, the first and second microphones MIC11 and MIC12 are respectively gated by first and second switch circuits.

The first switch circuit comprises a resistor R11, a resistor R12, a resistor R13, a capacitor C11 and a triode Q11, wherein a first end of the resistor R11 is connected with a base electrode of the triode Q11 and is connected with a power supply VCC through a resistor R12, and a second end of the resistor R11 is connected with a gating control signal GPIO 11; the emitter of the triode Q11 is grounded, and the collector is connected with the data output end of the first microphone; the second end of the resistor R13 is connected to the collector of the transistor Q11 and grounded via the capacitor C11, and the first end is used as the data output end of the first switch circuit.

The second switch circuit comprises a resistor R14, a resistor R15, a resistor R16, a capacitor C12 and a triode Q12, wherein the first end of the resistor R14 is connected with the base electrode of the triode Q12 and is grounded through a resistor R15, and the second end of the resistor R14 is connected with a gating control signal GPIO 12; the emitter of the triode Q12 is grounded, and the collector is connected with the data output end of the microphone; the second terminal of the resistor R16 is connected to the collector of the transistor Q12 and is grounded via the capacitor C12, and the first terminal thereof is used as the data output terminal of the second switch circuit.

Thus, the first microphone MIC11 and the second microphone MIC12 are gated by different gating control signals GPIO11 and GPIO12, respectively, sent by the main control chip. Specifically, the gating control signal GPIO11 sent by the main control chip controls the on/off of the switch of the first microphone MIC11, when the GPIO11 is at a high level, the transistor Q11 is turned on, and the data signal of the MIC11 is shorted to the ground, which is equivalent to that the MIC11 does not work. When the GPIO11 is at a low level, the transistor Q11 is turned off, the data signal of the MIC11 is output to the main control chip through the R13, and the MIC11 operates normally. The capacitor C11 is provided for absorbing switching noise of the first switching circuit.

The second microphone MIC12 is controlled by a gating control signal GPIO11 sent by the main control chip to switch on and off, when the GPIO12 is in a high level, the switch is opened, a data signal of the MIC11 is in short circuit with the ground, and the MIC11 does not work equivalently. When the GPIO12 is at low level, the Q12 is turned off, the data signal of the MIC12 is output to the main control chip through the R16, and the MIC12 operates normally. The capacitor C12 is provided for absorbing the switching noise of the second switching circuit.

In an initial state of the microphone circuit of the embodiment, when GPIOs 11 and 12 are in high impedance states, the first switch circuit is in an open state by default, a DATA signal of the MIC11 of the high AOP microphone is short-circuited to the ground, and DATA1 has no signal; the second switch circuit is in a closed state by default, the DATA signal of the MIC12 of the high SNR microphone is normally output, and the DATA2 signal is normal, so that the microphone circuit initial state is suitable for a common working scene by default.

It is understood that in other embodiments, the microphone circuit may be set in an initial state, the first circuit is turned off by default, the second circuit is turned on by default, and the DATA2 signal is output, so that the initial state of the microphone circuit is suitable for a noisy working scene by default.

Therefore, the gating circuit of the microphone array of the embodiment can realize switching between microphones with high SNR and high AOP, and combines the working environment of common working environment and particularly severe noise environment.

In a second exemplary embodiment of the present disclosure, a gating circuit is provided. The gating circuit is connected with at least one first electric appliance and at least one second electric appliance, a first parameter of the first electric appliance is within a first preset range, and a second parameter of the second electric appliance is within a second preset range. In this embodiment, the first electrical appliance is a first microphone, and the second electrical appliance is a second microphone. The first microphone MIC11 and the second microphone MIC12 are gated by a gating circuit.

In a similar manner to the first exemplary embodiment, the gating circuit is also a microphone data line gating circuit for gating the data line of the at least one first microphone and/or of the at least one second microphone. Unlike the first embodiment, the microphone data line gating circuit in this embodiment includes a first switching control circuit, a first switching circuit, and a second switching circuit.

The input end of the first switching control circuit is connected with a gating control signal; the first switching control circuit comprises a third on-off device which is connected to a second electrical appliance and gates a control signal;

the first switch circuit is connected with a first electric appliance, the input end of the first switch circuit is connected to the gating control signal, and the output end of the first switch circuit is a data output end of the first electric appliance; the first switch circuit comprises a first on-off device which is connected to a first electric appliance and gates a control signal;

the second switch circuit is connected with a second electrical appliance, the input end of the second switch circuit is connected to the first switch control circuit, the output end of the second switch circuit is a data output end of the second electrical appliance, the second switch circuit comprises a second on-off device, and the second on-off device is connected to the second electrical appliance and the first switch control circuit.

Fig. 2 is a circuit diagram of a data line gating circuit of a microphone according to a second embodiment of the present disclosure. As shown in fig. 2, the first microphone MIC11 is gated by a first switching circuit; the second microphone MIC12 is gated by the first switching control circuit and the second switching circuit.

The first switching control circuit comprises a resistor R22, a resistor R25, a triode Q23 and a capacitor C23, wherein the first end of the resistor R22 is connected with a power supply VCC, and the second end of the resistor R22 is connected with a gating control signal GPIO2 and the base electrode of the triode Q23 and is grounded through the capacitor C23; the emitter of the triode Q23 is grounded, and the collector is connected with a power supply VCC through a resistor R25;

the first switch circuit comprises a resistor R21, a resistor R23, a capacitor C21 and a triode Q21; the second end of the resistor R21 is connected with the base electrode of the triode Q23, and the first end of the resistor R21 is connected with the base electrode of the triode Q21; the emitter of the triode Q21 is grounded, and the collector is connected with the data output end of the first microphone; the second end of the resistor R23 is connected with the collector of the triode Q21 and is grounded through a capacitor C21, and the first end of the resistor R23 is used as the data output end of the first switch circuit;

the second switch circuit comprises a resistor R24, a triode Q22, a capacitor C22 and a resistor R26; the second end of the resistor R24 is connected with the collector of the triode Q23, and the first end is connected with the base of the triode Q22; the emitter of the triode Q22 is grounded, and the collector is connected with the data output end of the microphone; the second terminal of the resistor R6 is connected to the collector of the transistor Q22 and is grounded via the capacitor C22, and the first terminal thereof is used as the data output terminal of the second switch circuit.

Therefore, the first microphone MIC11 and the second microphone MIC12 realize one-out-of-two through the gating control signal GPIO2 sent by the main control chip. When the strobe control signal GPIO2 outputs a high level, the MIC12 outputs data to the main control chip; when the GPIO2 outputs a low level, the MIC11 outputs data to the main control chip.

In a third exemplary embodiment of the present disclosure, a gating circuit is provided. The gating circuit is connected with at least one first electric appliance and at least one second electric appliance, a first parameter of the first electric appliance is within a first preset range, and a second parameter of the second electric appliance is within a second preset range. In this embodiment, the first electrical appliance is a first microphone, and the second electrical appliance is a second microphone. The first microphone MIC11 and the second microphone MIC12 are gated by a gating circuit.

Different from the first and second embodiments, the gating circuit is a microphone power gating circuit, is used for gating the power supply of the at least one first electric appliance and/or the at least one second electric appliance, and comprises a second switching control circuit, a first power supply circuit and a second power supply circuit.

The input end of the second switching control circuit is connected with a gating control signal; the second switching control circuit comprises a first on-off device, and the first on-off device is connected with a gating control signal and a first power supply circuit;

the first power supply circuit is used for supplying power to a first electric appliance, the input end of the first power supply circuit is connected to the second switching control circuit, and the output end of the first power supply circuit is connected with the power supply input end of the first electric appliance; the first power supply circuit comprises a second on-off device, and the second on-off device is connected with the second switching control circuit and a power supply source of the first electric appliance;

the second power supply circuit is used for supplying power to a second electrical appliance, the input end of the second power supply circuit is connected to the gating control signal, and the output end of the second power supply circuit is connected with the power supply input end of the second electrical appliance; the second power supply circuit comprises a third on-off device, and the third on-off device is connected with the gating control signal and a second electrical appliance power supply.

Fig. 3 is a circuit configuration diagram of a power gating circuit of a microphone according to a third embodiment of the present disclosure. As shown in fig. 3, the second switching control circuit includes a resistor R31, a capacitor C31, a resistor R32, and a transistor Q31, wherein a first end of the resistor R31 is connected to the power VCC, a second end of the resistor R31 is connected to the gate control signal GPIO3, and is connected to the base of the transistor Q31 through a resistor R32, and is grounded through a capacitor C31; the emitter of transistor Q31 is connected to ground.

The first power supply circuit comprises a resistor R38, a resistor R39, a capacitor C34, a capacitor C35, a MOS transistor Q35, a resistor R33 and a triode Q32; the second end of the resistor R38 is connected with a power supply VCC, and the first end is connected with the collector of the triode Q32; the emitter of the triode Q32 is grounded, the base is connected with the collector of the triode Q31 through a resistor R33, and the collector is connected with the grid of the MOS transistor Q35 through a resistor R39; the source electrode of the MOS transistor Q35 is connected with a power supply VCC, the drain electrode is connected with the power supply input end of the first microphone and is grounded through a capacitor C35, a capacitor C34 is connected between the source electrode and the grid electrode of the MOS transistor Q35, and a parasitic diode is connected between the source electrode and the drain electrode.

The second power supply circuit comprises a resistor R35, a capacitor C32, a capacitor C33, an MOS transistor Q34, a resistor R36, a resistor R34 and a triode Q33; the second end of the resistor R35 is connected with a power supply VCC, and the first end is connected with the collector of the triode Q33; the emitter of the triode Q33 is grounded, the base is connected with the second end of the resistor R32 through the resistor R34, and the collector is connected with the gate of the MOS transistor Q34 through the resistor R36; the source electrode of the MOS transistor Q34 is connected with a power supply VCC, the drain electrode is connected with the power supply input end of the second microphone and is grounded through a capacitor C33, a capacitor C32 is connected between the source electrode and the grid electrode of the MOS transistor Q34, and a parasitic diode is connected between the source electrode and the drain electrode.

Therefore, the first microphone MIC11 and the second microphone MIC12 realize one-out-of-two through the gating control signal GPIO3 sent by the main control chip. When the GPIO3 outputs a high level, the power supply circuit of the MIC12 is turned on, and the power supply circuit of the MIC11 is turned off. The MIC12 works normally and outputs data to the main control chip; when the GPIO3 outputs a low level, the power supply circuit of the MIC11 is turned on, and the power supply circuit of the MIC12 is turned off. The MIC11 works normally and outputs data to the main control chip.

In a fourth exemplary embodiment of the present disclosure, a gating circuit is provided. The gating circuit is connected with at least one first electric appliance and at least one second electric appliance, a first parameter of the first electric appliance is within a first preset range, and a second parameter of the second electric appliance is within a second preset range. In this embodiment, the first electrical appliance is a first microphone, and the second electrical appliance is a second microphone. The first microphone MIC11 and the second microphone MIC12 are gated by a gating circuit.

In this embodiment, the gating circuit is configured to gate a power supply of the at least one first electrical appliance and/or the at least one second electrical appliance, and includes a first power supply circuit and a second power supply circuit.

The first power supply circuit is used for supplying power to a first electric appliance, the input end of the first power supply circuit is connected to a gating control signal GPIO41, and the output end of the first power supply circuit is connected with the power supply input end of the first electric appliance; the first power supply circuit comprises a first on-off device, and the first on-off device is connected with a gating signal and a first power supply of an electric appliance;

the second power supply circuit is used for supplying power to a second electrical appliance, the input end of the second power supply circuit is connected to the gating control signal GPIO42, and the output end of the second power supply circuit is connected with the power supply input end of the second electrical appliance; the second power supply circuit comprises a second on-off device, and the second on-off device is connected with the gating control signal and a second electrical appliance power supply.

Fig. 4 is a circuit configuration diagram of a power gating circuit of a microphone according to a third embodiment of the present disclosure. As shown in fig. 4, the first power supply circuit includes a resistor R41, a capacitor C41, a capacitor C42, a capacitor C43, a transistor Q41, a resistor R42, a resistor R43, and a MOS transistor Q42; the second end of the resistor R41 is connected with a gating signal GPIO41, and the first end is connected with the base electrode of the triode Q41; the second end of the resistor R42 is connected with a power supply VCC, and the first end is connected with the collector of the triode Q41; the emitter of the triode Q41 is grounded, the base is grounded through a capacitor C43, and the collector is connected with the gate of an MOS transistor Q42 through a resistor R43; the source electrode of the MOS transistor Q42 is connected with a power supply VCC, the drain electrode is connected with the power supply input end of the second microphone and is grounded through a capacitor C42, a capacitor C41 is connected between the source electrode and the grid electrode of the MOS transistor Q42, and a parasitic diode is connected between the source electrode and the drain electrode.

The second power supply circuit comprises a resistor R45, a capacitor C44, a capacitor C45, a capacitor C46, a triode Q43, a resistor R46, a resistor R47 and a MOS tube Q44; the second end of the resistor R45 is connected with a gating signal GPIO42, and the first end is connected with the base electrode of the triode Q43; the second end of the resistor R46 is connected with a power supply VCC, and the first end is connected with the collector of the triode Q43; the emitter of the triode Q43 is grounded, the base is grounded through a capacitor C46, and the collector is connected with the gate of an MOS transistor Q44 through a resistor R47; the source electrode of the MOS transistor Q44 is connected with a power supply VCC, the drain electrode is connected with the power supply input end of the second microphone and is grounded through a capacitor C45, a capacitor C44 is connected between the source electrode and the grid electrode of the MOS transistor Q44, and a parasitic diode is connected between the source electrode and the drain electrode.

Therefore, the first microphone MIC11 and the second microphone MIC12 realize alternative or simultaneous gating through the gating control signals GPIO41 and GPIO42 sent by the main control chip. When the GPIO41 outputs a high level, a power supply circuit of the MIC11 is turned on, the MIC11 works normally, and data are output to the main control chip; when the GPIO41 outputs a low level, a power supply circuit of the MIC11 is closed; when the GPIO42 outputs a high level, a power supply circuit of the MIC12 is turned on, the MIC12 works normally, and data are output to the main control chip; when the GPIO3 outputs a low level, the power supply circuit of the MIC11 is turned off.

In a fifth exemplary embodiment of the present disclosure, a microphone array circuit is provided. The microphone array circuit comprises a main control chip, an acquisition circuit unit and a gating circuit unit.

The master control chip is used for sending gating control signals of the microphone circuit and receiving first audio data of at least one first microphone and/or audio data of at least one second microphone.

The acquisition circuit unit comprises at least one first microphone and at least one second microphone, the AOP of the first microphone is higher than a first preset value, and the SNR of the second microphone is higher than a second preset value.

The gating circuit unit is connected to the gating control signal output end of the main control chip, the gating circuit unit can be a microphone data line gating circuit or a microphone power supply gating circuit, and the microphone data line gating circuit can adopt the gating circuits described in the first embodiment and the second embodiment as an example; alternatively, the microphone power gating circuit may adopt the gating circuit as described in the third and fourth embodiments.

It is understood that the gating circuit unit may also be a multi-port input unit, an input port of the multi-port input unit is connected to the data output ports of the at least one first microphone and the at least one second microphone, and an output port of the multi-port input unit is connected to the main control chip. The method requires that the main control chip has enough microphone data input channels, all the microphone data are input to the main control chip, and the main control chip actively selects the microphone data to be used.

Therefore, through the microphone array circuit of the embodiment, switching between microphones with high SNR and high AOP can be realized through different working scenes, and the working environment of a common working environment and a particularly severe noise environment is considered.

In a sixth exemplary embodiment of the present disclosure, there is also provided an acoustic enclosure including the microphone array circuit according to the fourth exemplary embodiment. The acquisition circuit unit comprises a plurality of acquisition subunits, each acquisition subunit comprises at least one first microphone and at least one second microphone, and the acquisition subunits are uniformly distributed and mounted on a carrier, wherein the carrier can be a circuit board or a sound box shell. By the distribution mode, the sounds received in different directions can be more real, and different microphones can be gated more accurately. For example, referring to fig. 1A again, each of the collecting subunits includes a first microphone and a second microphone, and the collecting subunits are uniformly distributed around the circumference of the enclosure.

In this embodiment, the first microphone MIC11 is a high AOP microphone, the second microphone MIC12 is a high SNR microphone, the MIC21 is a high AOP microphone, the MIC22 is a high SNR microphone, and so on. The high AOP microphone type numbers comprise SPH0655LM4H-1, SPW0690LM4H-1 and the like; high SNR microphone models include BOM9736RL-T, MD-DRA361-P10 and the like.

Furthermore, the sound box can be further provided with a sound intensity monitoring device, the sound intensity monitoring device is connected with the main control chip and used for monitoring the external sound intensity and sending the monitored sound intensity data to the main control chip, and the main control chip enables the microphone to be gated according to the sound intensity data. Wherein, the sound intensity monitoring device can be a sound intensity monitor, a sound intensity monitoring circuit and the like. The other scheme is as follows: the main control chip has a sound intensity detection function, can acquire sound signals collected by a microphone in the collection circuit unit, and detects the sound signals to obtain sound intensity data. When any one microphone in the acquisition circuit unit works, the main control chip can acquire external sound intensity data, namely, a pickup sound intensity level according to the microphone. The pickup sound intensity level is used for comparing with a preset switching threshold value to decide which microphone data is used. Both of the above approaches are applicable to the present disclosure.

In a seventh exemplary embodiment of the present disclosure, a speech recognition method is provided. Fig. 5 is a flow chart of the disclosed speech recognition method. As shown in fig. 5, the voice recognition method includes:

s101, sound intensity data of pickup is monitored, and work scene judgment is carried out according to the sound intensity data of pickup to obtain a work scene judgment result;

s102, selecting first audio data of at least one first microphone or second audio data of at least one second microphone according to the working scene judgment result, wherein a first parameter of the first microphone is within a first preset range, and a second parameter of the second microphone is within a second preset range;

and S103, performing voice recognition by using the selected first audio data or second audio data.

Wherein the step S101 further includes:

Illustratively, the first parameter of the first microphone is AOP, which is higher than a first preset value, and the second parameter of the second microphone is SNR, which is higher than a second preset value. When the picked sound intensity data is larger than the preset switching threshold value, selecting first audio data output by a first microphone for voice recognition; and when the pickup sound intensity data is less than or equal to the preset switching threshold value, selecting second audio data output by a second microphone for voice recognition.

And default of the initial value of the working scene judgment result is a second working scene applicable to the second microphone or a first working scene applicable to the first microphone. Illustratively, in the microphone circuit in the initial state, the audio data signal of the high-SNR microphone is normally output, so that the initial state of the microphone circuit is applicable to a common working scene by default; the microphone circuit can be set in an initial state, and audio data signals of the high-AOP microphone are normally output, so that the microphone circuit is suitable for a noisy working scene in the initial state by default.

In step S102, the gating control signal may be a data line gating control signal and/or a power supply gating control signal.

Specifically, the gating circuits according to the first, second, third, and fourth embodiments may be adopted, and only one of the picked-up sound data of the two sets of microphones (high AOP or high SNR) is in an operating state at the same time. The step S102 may include:

In a specific embodiment, scene judgment and working mode switching are carried out before voice processing of the module is awakened each time, mode switching is not carried out in a single voice interaction process, and an awakening-recognition-response process is completed once. The process flow is shown in fig. 6.

Furthermore, if a certain working mode is selected in the single voice interaction recognition process, the environment suddenly changes violently, so that normal recognition cannot be performed, the gating control signal can be sent again, and data of another group of microphones can be gated for recognition processing. The process flow is shown in fig. 7.

In other embodiments, the gating circuit of the first embodiment may be further adapted to gate at least one first microphone and at least one second microphone at the same time, or a multi-port input unit may be further adapted, where an input port of the multi-port input unit is connected to data output ports of the at least one first microphone and the at least one second microphone, and the input port of the multi-port input unit sends picked-up sound data of the two-mode microphones to the main control chip at the same time. In this case, the main control chip may select one of the audio data of the first microphone and the second microphone to perform voice recognition, or select and combine two audio data to perform voice recognition according to the requirement.

Specifically, the step S102 may further include:

transmitting a gating control signal to gate the at least one first microphone and the at least one second microphone, and simultaneously buffering and storing first audio data of the at least one first microphone and second audio data of the at least one second microphone;

Further, after storing the first audio data and the second audio data of the two-mode microphone, the main control chip may select the first audio data or the second audio data for voice recognition according to the working scene determination result, obtain a voice recognition result, and determine whether the voice recognition result is correct, and if the voice recognition result is correct, take the result as a final voice recognition result; otherwise, the other group of audio data is adopted for voice recognition, and the two voice recognition processing results are synthesized for judgment to obtain the final voice recognition result. The process flow is shown in fig. 8.

For example, in a situation where the voice command cannot be correctly recognized in one operation mode, or only a part of the command can be correctly recognized, such as: "i want to listen to the whitish balloon of zhou-jeren", one mode may not recognize the instruction with the correct meaning at all, or may recognize only a portion of "i want to listen to zhou-jeren … …". At the moment, the auxiliary use of the pickup data of the other mode is carried out, and then comprehensive judgment is carried out to obtain a correct voice command. For example, when a voice recognition device in a kitchen, such as a range hood, an electric cooker, etc., has an intelligent voice recognition function, the noise environment is complicated and greatly different at different times due to the fact that the noise in the kitchen is serious, such as the sound of the range hood, the sound of cooking, etc. Under the condition, the microphone audio data in any single mode cannot obtain a good identification effect, so that a high-AOP and high-SNR dual-microphone type microphone array is adopted, two paths of microphone data are simultaneously transmitted to a main control chip, the main control chip simultaneously processes the two paths of data, one path of microphone is used for listening to loud sound, the other path of microphone is used for listening to details, a final result is obtained, and the identification effect of a product is improved.

Further, the step S103 further includes: voice wake-up is performed prior to voice recognition using the selected first audio data or second audio data.

In an eighth exemplary embodiment of the present disclosure, a speech recognition apparatus is provided. Fig. 9 is a schematic structural diagram of a speech recognition apparatus according to an embodiment of the disclosure. As shown in fig. 9, the speech recognition apparatus includes a mode selection preprocessing unit, a working mode selection unit, and a speech recognition processing unit.

The mode selection preprocessing unit is used for monitoring pickup sound intensity data and judging a working scene according to the pickup sound intensity data to obtain a working scene judgment result; the working mode selection unit is connected to the mode selection preprocessing unit and used for selecting first audio data of at least one first microphone or second audio data of at least one second microphone according to the working scene judgment result, wherein the AOP of the first microphone is higher than a first preset value, and the SNR of the second microphone is higher than a second preset value; the voice recognition processing unit is connected to the working mode selection unit and used for carrying out voice recognition on the data of the gated microphone.

Further, the mode selection preprocessing unit further includes a comparing unit, configured to compare the monitored pickup sound intensity data with a preset switching threshold, and if the pickup sound intensity data is greater than the preset switching threshold, determine that the current scene is a first working scene suitable for the first microphone; and if the pickup sound intensity data is smaller than or equal to the preset switching threshold value, judging that the current scene is a second working scene suitable for the second microphone.

Further, the voice recognition processing unit further comprises a voice wake-up unit, configured to perform voice wake-up before recognition.

In an eighth exemplary embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the speech recognition method according to the sixth embodiment.

For the purpose of brevity, any technical features that can be applied to the same embodiment are described herein, and the same description need not be repeated.

So far, the embodiments of the present disclosure have been described in detail with reference to the accompanying drawings. It is to be noted that, in the attached drawings or in the description, the implementation modes not shown or described are all the modes known by the ordinary skilled person in the field of technology, and are not described in detail. Further, the above definitions of the various elements and methods are not limited to the various specific structures, shapes or arrangements of parts mentioned in the examples, which may be easily modified or substituted by those of ordinary skill in the art.

Furthermore, the word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements.

The use of ordinal numbers such as "first," "second," "third," etc., in the specification and claims to modify a corresponding element does not by itself connote any ordinal number of the element or any ordering of one element from another or the order of manufacture, and the use of the ordinal numbers is only used to distinguish one element having a certain name from another element having a same name.

In addition, unless steps are specifically described or must occur in sequence, the order of the steps is not limited to that listed above and may be changed or rearranged as desired by the desired design. The embodiments described above may be mixed and matched with each other or with other embodiments based on design and reliability considerations, i.e., technical features in different embodiments may be freely combined to form further embodiments.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, this disclosure is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the present disclosure as described herein, and any descriptions above of specific languages are provided for disclosure of enablement and best mode of the present disclosure.

The disclosure may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. Various component embodiments of the disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components in the relevant apparatus according to embodiments of the present disclosure. The present disclosure may also be embodied as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present disclosure may be stored on a computer-readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Also in the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the disclosure, various features of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various disclosed aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, the claimed disclosure requires more features than are expressly recited in each claim. Rather, as the following claims reflect, disclosed aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this disclosure.

The above-mentioned embodiments are intended to illustrate the objects, aspects and advantages of the present disclosure in further detail, and it should be understood that the above-mentioned embodiments are only illustrative of the present disclosure and are not intended to limit the present disclosure, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. A speech recognition method, comprising:

2. The speech recognition method of claim 1, wherein the first parameter AOP of the first microphone is higher than a first preset value and the second parameter SNR of the second microphone is higher than a second preset value.

3. The speech recognition method of claim 1, wherein the selecting the first audio data of the at least one first microphone or the second audio data of the at least one second microphone comprises:

4. The speech recognition method according to claim 1, wherein the performing a work scene determination according to the picked-up sound intensity data to obtain a work scene determination result includes:

5. The speech recognition method of claim 4, wherein the selecting the first audio data of the at least one first microphone or the second audio data of the at least one second microphone according to the working scenario determination result comprises:

6. The speech recognition method of claim 4, wherein the selecting the first audio data of the at least one first microphone or the second audio data of the at least one second microphone according to the working scenario determination comprises:

7. The speech recognition method of claim 6, further comprising:

selecting the first audio data or the second audio data to perform voice recognition according to the working scene judgment result, judging whether the voice recognition result is correct or not after the voice recognition result is obtained, and taking the voice recognition result as a final voice recognition result if the voice recognition result is the correct result; otherwise, the other group of audio data is adopted for voice recognition, and the two voice recognition processing results are synthesized for judgment to obtain the final voice recognition result.

8. The speech recognition method of claim 4, further comprising:

9. A speech recognition apparatus, comprising:

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the speech recognition method according to any one of claims 1 to 8.